Is there any improvement in simulation of wintertime Western Pacific teleconnection pattern and associated climate anomalies in CMIP6 comparing with CMIP5 models?

The western Pacific pattern (WP) is one of the most important atmospheric teleconnections over the Northern Hemisphere (NH) in boreal winter, which plays key roles in regulating weather and climate variations over many parts of the NH. This study evaluates ability of the coupled models participated in CMIP5 and CMIP6 in capturing the spatial pattern, dominant frequency, and associated climate anomalies of the winter WP. Ensemble means of the CMIP5 and CMIP6 models well capture spatial structures of the WP, with slightly higher skills for the CMIP6. However, the northern (southern) centre of the WP is shifted westward (eastward) relative to the observations, and the strength of the northern centre is overestimated in most CMIP5 and CMIP6 models. CMIP6 shows an improvement in simulating the dominant periodicity of the WP. WP-related climatic anomalies in most parts of the NH can be well simulated. However, there exists a large spread across the models in simulating surface air temperature (SAT) anomalies in Russian Far East and Northwest North America, which is attributable to the diversity of the intensity of the WP’s northern lobe. Most CMIP5 and CMIP6 models largely overestimate the WP-related precipitation anomalies over Siberia, which is partly due to the overestimation of mean precipitation there. Furthermore, most models simulate a close relation of the WP and Arctic Oscillation (AO), which does not exist in observation. The CMIP5 and CMIP6 models with weak WP-AO relations have better ability than the models with strong WP-AO relations in capturing the WP-related SAT and precipitation anomalies over the NH, especially over Eurasia.